Separate realities; a plain narrative of A'posteriori cognition: an analogue for comparisons with and between Asperger's syndrome and other autistic spectrum conditions.

Andrew Walker

1997, Living and Learning with Autism: Perspectives from the Individual, the Family and the Professional. Sunderland, England: Autism Research Unit. pp 19-27.


A model is proposed to describe certain common cognitive characteristics of autistic spectrum group individuals. The A'posteriori model is apparently wholly concordant with seldom cited models developed independently by other autistic authors. The model suggests the necessity for a reappraisal of diagnostic criteria and a rationalisation of the prevailing taxonomic classifications of the autistic spectrum conditions based upon a clear and common cognitive familial link.


There are references to autism and a detachment from the social world (Frith1991) as early 1916 which were effectively forgotten until rediscovered by later workers in 1940s. consensus of opinion appeared then be that was essentially communication disability. Succeeding generations researchers have added their own work result is these taxonomic embellishments caused spawn family its own. well now Asperger syndrome (AS)schizoid personality disorder (SPD)Kannerhigh-functioning an extended associated conditions. considerable academic debate surrounding classification parameters so-called related conditions.

A further complicating factor is that the definitions and parameters of AS and associated disorders show a history of modification and continuing change. For example, Volkmar et al. (1985) suggest that AS is not distinct from Autism whereas Wolff (1989), Wolff and Cull (1986) and Wolff and McGuire (1995) consistently use the term SPD interchangeably with AS, stating clearly that they consider them to be literally one and the same condition. In contrast Tantam (e.g. 1988, 1991) believes that AS and SPD are clearly distinct, whilst other workers such as Bishop (1989) explore other AS "borderlands" such as that with semantic-pragmatic disorder, whilst also considering the appropriateness of parent categories of AS such as that of Pervasive Development Disorder (PDD). Happé and Frith (1991) go further and suggests that the PDD association be eliminated altogether with respect to autism and proposes it's substitution with one of "Autism Spectrum Disorders".

An agreed scientific definition of autism therefore does not yet apparently exist. This undefined condition is currently delineated from conditions which are apparently un-autistic by the observation and interpretation of certain apparently associated behaviours, the cause and purpose of which are apparently most often not understood. The principal current diagnostic tools available are DSM-IV (APA, 1995) or ICD-10 (WHO, 1993) though in practice interpretation appears to vary widely. Autistic people have also produced their own work on the subject independent of the scientific establishment. Though popular with the autistic, their carers and the general reader, this material is of less interest to the researcher, presumably because the biographical format is largely anecdotal (e.g. Williams, 1992) and the content is thus unquantifiable.

Significantly, Asperger and other early researchers did not appear to attempt to delineate sub-categories of the autistic condition. They may have appreciated, as many autistic people appear to know intuitively, that there appears to be a common cognitive system. The apparent variation between autistic individuals from various points on the autistic continuum may be due entirely to natural variation between personalities and/or unrecognised secondary disorders.

A computer metaphor is useful to define the parameters of the cognitive system; for instance the hardware, operating system and software of a computer may be substituted for the physical brain, the cognitive system, and personality respectively. Autism is based firmly in the hardware and as such is fixed and unalterable. It is distinct from, but is expressed through the personality via the operating, or cognitive, system which because it is instinctive, is again genetically predetermined. As it is impractical to study directly the physiological processes of the autistic brain, examination is confined to the distinctive characteristic autistic cognitive system. Like an operating system, the autistic cognitive system is dedicated to a particular brand of hardware, in this case the autistic type. However, there is considerable variation between the personalities of autistic people and each is unique. It is therefore essential to find a reliable way to distinguish the developed personality, the evolved software, from the underlying operating, or cognitive system.

To help clarify the distinction between autism associated cognition and personality, if the range of about one hundred thousand possible trim and paint options of a single brand of truck represents the range of variation in the apparent personalities of autistic people, and if a certain aspect of appearance such as colour is selected arbitrarily to define brands of truck, then there would be as many brands of truck as there are colours. However, if one were to categorise by chassis then they are all found to be of a single type. It is suggested that if many of the autistic sub-types were to be redefined by cognitive system, then there may only be a single type.

The a'posteriori cognitive model

This model describes only the cognitive system as I have already defined it. It compares the respective instinctive subconscious processes of autistic and non-autistic people and demonstrates how the former may be contrived to survive social situations by generating a conscious but effectively synthetic persona to fit certain social situations. This 'A posteriori' process, so-called because of the apparently logical step-by-step approach, may be considered as being analogous to some of the fundamental functions of elementary computer programming language. Decision making lines of code such as the 'if-then-else' statement, loops, and the nesting of loops appear to be similar in concept to the linear or serial nature of the apparent processing implicit in the A'posteriori model.

In order to contextualise the A'posteriori model as shown at Figure 1, which is an analogue for the autistic condition, a baseline comparison is required with the non-autistic condition; namely the 'Allegorical' model shown at Figure 2. The entire shaded area of the Allegorical model represents 'self' whilst the heaviest shading shows the unavoidable permeable interface which divides self from the external environment at left; this is essentially a two way filter which lends a social context or interpretation to every observation or response.

Principal cognitive elements are embedded in this cognitive environment. They are the respective locations for the principal cognitive functions of accessing experience, strictly logical functions and the processing of observed social behaviour and responses. These social messages, inherently heavy with symbolism, are effectively algorithms. This latter element is the dominant function in the model, is dynamic and is a 'learning' system. It is significant that logical processing is at a specific location and is apparently isolated. There are interactive processes between the cognitive elements. It is essential to note that any information passing into the filter at any time and for whatever reason, is automatically assigned a label of social meaning.

When considering the cognitive environment of the A'posteriori model, the essential difference to the Allegorical model is the almost complete absence of a social interface and that there are two 'zones' within the A'posteriori cognitive environment. The logical processing zone, though different in function and purpose to the social interface of the Allegorical model, occupies an approximately equivalent location in the system. It is important to note that the logical processing is not at a specific site but is apparently extensive. The active principal elements, those which manipulate and process a given stimulus are wholly within a zone of conscious behaviour of logical reasoning. Experience is shown to reside in an area characterised as storage, where no apparent conscious processing occurs. To extend the computer metaphor, the logical processing and process 'zones' may be considered to be analogous to the random access memory (RAM), and the hard disk, of a personal computer respectively. Though reference is made to zones and elements, this does not imply actual locations but rather a context in which the stimulus data are manipulated. There are no principal cognitive elements in the A'posteriori model which are equivalent in function to the compilation element in the social filter of the Allegorical model. The nearest possible equivalent is the combination of interpretation and response elements, though significantly they are located within the logical zone.

The Allegorical model recognises several distinct stimuli in the external environment, can comprehend their meaning and respond in appropriate terms to each concurrently. This is parallel processing and occurs wholly within the filter and is apparently wholly instinctive. For instance, this allows an individual to talk to someone whilst at the same time monitor and respond to eye contact, body language and many other cues at the same time.

The A'posteriori model can focus only upon one source of information at any one time. This is serial processing. An autistic person has therefore to make a decision about what they should concentrate upon or risk being overwhelmed with too much information. For example, an autistic individual may choose to concentrate upon the literal verbal content of a conversation but ignore all else to produce a flat apparently emotionless but factually correct android-like response. Then again the individual may concentrate upon another aspect of the interaction, such as the other individual's body posture, and completely miss the literal meaning and context of the conversation. Increase the number of individuals, namely variables, involved and the difficulties are compounded.

However, there is a partial though contrived solution to this problem. Shown as unshaded areas and superimposed upon the model, are the essential components of the learning acquired social skills mechanism. Though there can be no change to the basic cognitive system or the serial input, this is a conscious process whose purpose is to broaden the nature of the output; in effect to make the 'android' smile as well as talk. However, because it requires considerable concentration there is a high cost in terms of conscious effort; analogous perhaps to the effort required to converse in a foreign language before the point is reached when an individual may also 'think' in that language.


There exists a remarkably similar and infamous example (Bakker, 1986) to that of autism of the process of evolution of the understanding of a subject. In the mid-nineteenth century, palaeontologists assigned all of the dinosaurs and birds to their own phylogenetic group. However, since then workers have reclassified and sub-divided them as reptiles for reasons apparently no better than that dinosaurs appear similar to reptiles, if only superficially. It has been proven recently that the original conclusions of the nineteenth century palaeontologists was, in general terms, completely correct. By the same token, were Kanner and Asperger correct in believing that autism is a singular entity? Some current researchers now appear to support this, at least in part.

For instance, Wolff (Wolff, 1995) has noted major clinical similarities between SPD and AS diagnosed individuals, and perhaps also high-functioning autism and Kanner's, in subjects from both her own and from other studies. Wolff accounts for the differing diagnoses by noting that the respective authors had used 'differing but overlapping diagnostic criteria' and states that: "At the present time, therefore, the term 'Asperger's syndrome', as defined in ICD-10 and as used by Wing, Tantam and also by Frith and her colleagues in their experimental work (1991), seem to imply a condition which may be equivalent to high-functioning autism and which is more serious than that of the children described by both Asperger and ourselves...". In an effort towards a taxonomic rationalisation, Wolff goes on to propose 'schizoid/Asperger disorder' as a new classificatory label.

This example serves to highlight the difficulties and anxiety caused to the affected individual when attempting to obtain an appropriate diagnosis, and would appear to suggest that the nature of the individual's diagnosis may actually depend to an extent upon which person is consulted and their personal 'interpretation' of ICD-10 or DSM-V. For instance, for an individual with an existing diagnosis of AS, then the SPD label may be equally or more applicable according to Wolff et al, or else the high-functioning autistic label may be more appropriate according to Volkmar et al. (1985).

A fact apparently unrecognised by researchers in autism is that a number of autistic people, with varying formal diagnoses and differing national and cultural backgrounds, have independently developed cognitive models significantly similar to the A'posteriori model (e.g. Madar, 1996; Williams, 1996).


It is suggested that there is essentially only a single clinical entity which is autism when using the cognitive system as the sole classificatory criterion. This system is apparently serial in process, as opposed to parallel in the non-autistic condition, and is the source of the characteristic autistic social impairment. The corollary is that, though autistic people are born apparently similar, outcomes may be largely dependent upon environmental factors.


American Psychiatric Association (1995) Diagnostic and statistical manual of mental disorders (4th ed.) Washington DC: American Psychiatric Association.

Bakker, R. (1986) The Dinosaur Heresies. Harlow, Essex: Longman Scientific & Technical.

Bishop, D. V. (1989) Autism Asperger's syndrome and semantic-pragmatic disorder: Where are the boundaries?, British Journal of Disorders of Communication, 24, 2, 107-121.

Frith, U. (1991) Asperger and his syndrome. In: Frith, U. (ed.) Autism and Asperger syndrome. Cambridge: Cambridge University Press.

Happé, F. G. E. and Frith, U. (1991) Is autism a pervasive developmental disorder? Debate and argument: How useful is the "PDD" label?, Journal of Child Psychology and Psychiatry and Allied Disciplines, 32, 7, 1167-1168.

Madar, T. (1996) The Primary Characteristics of Asperger Syndrome, Asperger United, 9, 3-11.

Tantam, D. (1988) Lifelong isolation and eccentricity: II. Asperger's syndrome or Schizoid Personality Disorder? British Journal of Psychiatry, 153, 783-791.

Tantam, D. (1991) Asperger syndrome in adulthood. In: Frith, U. (ed.) Autism and Asperger syndrome. Cambridge: Cambridge University Press.

Volkmar, F. R., Paul, R., and Cohen, D. J. (1985) The use of "Asperger's syndrome", Journal of Autism and Developmental Disorders, 15, 4, 437-439.

Williams, D. (1992) Nobody Nowhere. London: Doubleday.

Williams, D. (1996) Autism: An Inside-Out Approach. London: Jessica Kingsley.

Wolff, S. (1989) Constitutional aspects of personality beginning in childhood: Schizoid personality disorder (Asperger's syndrome) In: Doxiadis, S. and Stewart, S. (eds.) Early influences in shaping the individual. NATO Advanced Science Institutes series. Series A: Life sciences, Vol. 160. New York: Plenum Press.

Wolff, S. and Cull, A. (1986) "Schizoid" personality and antisocial conduct: A retrospective case note study, Psychological Medicine, 16, 3, 677-687.

Wolff, S. (1995) Loners: The life path of unusual children. London: Routledge.

Wolff, S. and McGuire, R. J. (1995) Schizoid personality in girls: A follow-up study: What are the links with Asperger's syndrome?, Journal of Child Psychology and Psychiatry and Allied Disciplines, 36, 5, 793-817.

World Health Organisation (1993) The ICD-10 Classification of Mental and Behavioural Disorders: Diagnostic criteria for research. Geneva: World Health Organisation.


Related articles:

Andrew Walker: What is the point of autism?

Dinah Murray, Mike Lesser and Wendy Lawson: Monotropism Hypothesis

Mike Lesser and Dinah Murray: Mind as a Dynamical System: Implications for Autism

Mike Lesser and Dinah Murray: A model of the Interest System

Mike Lesser and Fergus Murray: NOI4 - Interactive Animation of the Interest System

Zurück nach oben